JP2020038374A - Audio encoder for encoding multichannel signal and audio decoder for decoding encoded audio signal - Google Patents

Abstract

To provide an audio encoder for encoding a multichannel audio signal.SOLUTION: An audio encoder 2 includes a linear prediction domain encoder 6, a frequency domain encoder 8, and a controller 10 for switching between the linear prediction domain encoder 6 and the frequency domain encoder 8. The controller presents a portion of the multichannel signal either by an encoded frame of the linear prediction domain encoder or by an encoded frame of the frequency domain encoder. The linear prediction domain encoder includes a downmixer 12 for downmixing the multichannel signal 4 to obtain a downmixed signal 14, and a linear prediction domain core encoder 16 for encoding the downmix signal. The linear prediction domain encoder includes a first joint multichannel processor 18 for generating first multichannel information 20 from the multichannel signal 4.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an audio encoder for encoding a multi-channel audio signal and an audio decoder for decoding an encoded audio signal. Embodiments relate to a switched perceptual audio coder that includes waveform preservation and parametric stereo coding.

  The perceptual encoding of audio signals for the purpose of data reduction for efficient storage or transmission of these signals is a widely used practice. In particular, when the highest efficiencies need to be achieved, encoders are used that closely adapt to the signal input characteristics. One example is ACELP (Algebraic Code-Excited Linear Prediction) encoding of speech signals, TCX (Transform Coded Excitation) of background noise and mix signals, and transform coding excitation of music content. An MPEG-D USAC core encoder that can be configured to mainly use AAC (Advanced Audio Coding). All three inner encoder configurations are switched instantaneously in a signal adaptation manner corresponding to the content of the signal.

  In addition, combined multi-channel coding techniques (such as intermediate / side coding) or, for highest efficiency, parametric coding techniques are used. Parametric coding techniques basically aim at recreating a perceived equivalent audio signal, rather than a faithful reconstruction of a given waveform. Examples include noise filling, bandwidth extension, and spatial audio coding.

  When combining the signal-adaptive core coder with either one of the state-of-the-art encoder combined multi-channel or parametric coding techniques, the core encoder is switched to match the signal characteristics, but the M / S The choice of multi-channel coding technology, such as stereo, spatial audio coding or parametric stereo, remains fixed and independent of signal characteristics. These techniques are typically used for the core encoder and for the core encoder as a pre-processor and for the core decoder as a post-processor (both without knowing the actual choice of core encoder).

  On the other hand, the choice of a parametric coding technique for bandwidth extension sometimes depends on the signal. For example, techniques applied in the time domain are more efficient for speech signals, while frequency domain processing is more relevant for other signals. In such cases, the multi-channel coding technique employed must be compatible with both types of bandwidth extension techniques.

Related topics of the state of the art include:
PS and MPS as front / back processor for MPEG-D USAC core encoder
MPEG-D USAC standard MPEG-H 3D audio standard

ISO / IEC DIS23003-3, Usac ISO / IEC DIS23008-3, 3D audio

  In the MPEG-D USAC, a switchable core coder is described. However, in the USAC, the multi-channel coding technique is a fixed fixed common to the entire core coder, independent of its internal switch of coding principle being ACELP or TCX ("LPD") or AAC ("FD"). Defined as a choice. Thus, if a switched core encoder configuration is required, the encoder is limited to using parametric multi-channel coding (PS) for the entire signal. However, for example for music signal coding, it switches rather dynamically between L / R (left / right) and M / S (middle / side) schemes per frequency band and per frame It is more appropriate to use a combined stereo coding that can.

  Thus, there is a need for an improved approach.

  It is an object of the present invention to provide an improved concept for processing an audio signal. This object is solved by the subject matter of the independent claims.

  The present invention is based on the discovery that a (time-domain) parametric encoder using a multi-channel coder is advantageous for parametric multi-channel audio coding. The multi-channel coder may be a multi-channel residual coder that reduces the bandwidth for transmitting coding parameters as compared to individual coding for each channel. For example, it is advantageously used in combination with a frequency domain coupled multi-channel audio coder. The time domain combined multi-channel coding technique and the frequency domain combined multi-channel coding technique are combined so that, for example, a frame-based decision can guide the current frame to a time-based or frequency-based coding period. . That is, embodiments provide combined multi-channel coding and parametric spatial audio coding in a fully switchable perceptual coder that allows the use of different multi-channel coding techniques depending on the choice of core coder. 4 illustrates an improved concept for combining switchable core encoders using optimization. This is advantageous because, in contrast to existing methods, the embodiments show a multi-channel coding technique that is immediately switched to the core coder at the same time, and therefore adapts closely to the choice of the core coder. Thus, the described problems that appear due to the fixed choice of multi-channel coding technique are avoided. Furthermore, a fully switchable combination of a given core coder and the associated multi-channel coding technique is possible. Coders such as AAC (Advanced Audio Coding) that use L / R or M / S stereo coding, for example, use dedicated joint stereo or multi-channel coding, eg, frequency domain (FD) using M / S stereo. 2.) There is a possibility that the music signal is encoded in the core coder. This decision is applied separately for each frequency band within each audio frame. For example, in the case of speech signals, the core coder immediately switches to a linear predictive decoding (LPD) core coder and its associated different, eg, parametric, stereo coding techniques.

  The embodiments show a stereo processing based solely on the mono LPD path and a stereo signal based seamless switching scheme combining the output of the stereo FD path with the output from the LPD core coder and its dedicated stereo coding. This is advantageous because it allows switching of seamless encoders without artifacts.

  Embodiments relate to an encoder for encoding a multi-channel signal. The encoder includes a linear prediction domain encoder and a frequency domain encoder. Further, the encoder includes a controller for switching between a linear prediction domain encoder and a frequency domain encoder. The linear prediction domain encoder further includes a downmixer for downmixing the multi-channel signal to obtain a downmix signal, a linear prediction domain core encoder for encoding the downmix signal, and a first multi-channel signal from the multi-channel signal. A first multi-channel encoder for generating channel information; The frequency domain encoder includes a second combined multi-channel encoder for encoding second multi-channel information from the multi-channel signal. The second multi-channel encoder is different from the first multi-channel encoder. The controller is configured such that a portion of the multi-channel signal is represented by either an encoded frame of a linear prediction domain encoder or an encoded frame of a frequency domain encoder. The linear prediction domain encoder includes an ACELP core encoder and, for example, a parametric stereo coding algorithm as a first combined multi-channel encoder. The frequency domain encoder includes, for example, an AAC core encoder using, for example, L / R or M / S processing as the second combined multi-channel encoder. The controller analyzes the multi-channel signal for frame characteristics, such as speech or music, and determines either an individual frame, a series of frames or portions of the multi-channel audio signal, either a linear prediction domain encoder or a frequency domain encoder. Is used to encode this portion of the multi-channel audio signal.

  The embodiment further shows an audio decoder for decoding the encoded audio signal. The audio decoder includes a linear prediction domain decoder and a frequency domain decoder. The audio decoder further includes a first combined multi-channel decoder for generating a first multi-channel representation using the output of the linear prediction domain decoder and the multi-channel information, and an output of the frequency domain decoder and the second multi-channel information. And a second multi-channel decoder for generating a second multi-channel representation using the same. Further, the audio decoder includes a first combiner for combining the first multi-channel representation and the second multi-channel representation to obtain a decoded audio signal. The combiner performs a seamless, artifact-free switch between a first multi-channel representation, eg, a linear prediction multi-channel audio signal, and a second multi-channel representation, eg, a frequency domain decoded multi-channel audio signal. I do.

  The embodiment shows a combination of ACELP / TCX coding in the LPD path with dedicated stereo coding and independent AAC stereo coding of the frequency domain path in the switchable audio coder. In addition, embodiments show seamless instantaneous switching between LPD and FD stereo. Another embodiment relates to independent selection of combined multi-channel coding for different signal content types. For example, parametric stereo is used for speech that is primarily encoded using the LPD path. On the other hand, more adaptive stereo encoding is used for music encoded in the FD path. It can dynamically switch between L / R and M / S schemes on a per frequency band and per frame basis.

  According to an embodiment, for speech that is mainly encoded using the LPD path and is always centered in the stereo image, simple parametric stereo is appropriate. On the other hand, music encoded in the FD pass always has a more sophisticated spatial distribution and can benefit from more adaptive stereo coding. It can dynamically switch between L / R and M / S schemes on a per frequency band and per frame basis.

  Another embodiment includes a downmixer (12) for downmixing the multichannel signal to obtain a downmix signal, a linear prediction domain core encoder for encoding the downmix signal, and a spectrum of the multichannel signal. FIG. 2 shows an audio encoder including a filter bank for generating a representation and a combined multi-channel encoder for generating multi-channel information from a multi-channel signal. The downmix signal has a low band and a high band. The linear prediction domain core encoder is configured to apply a bandwidth extension process to parametrically encode the high band. Further, the multi-channel encoder is configured to process a spectral representation of the multi-channel signal that includes a low band and a high band. This is advantageous because individual parametric coding can use its optimal time-frequency decomposition for obtaining its parameters. This is implemented, for example, using a combination of ACELP (Algebraic Code Excited Linear Prediction) + TDBWE (Time Domain Bandwidth Extension). ACELP encodes the low band of the audio signal, TDBWE encodes the high band of the audio signal, and encodes parametric multi-channel encoding with an external filter bank (eg, DFT). This combination is particularly efficient because it is known that the best bandwidth extension for speech is in the time domain and that multi-channel processing should be in the frequency domain. Since ACELP + TDBWE does not have any time-frequency converter, an external filter bank or transform such as DFT is advantageous. Further, the framing of the multi-channel processor is the same as that used in ACELP. Even if the multi-channel processing is done in the frequency domain, the temporal resolution for its parameter calculation or downmix is ideally close to or even equal to ACELP framing.

  The described embodiments are advantageous because independent choices of combined multi-channel coding apply for different signal content types.

  Embodiments of the present invention will be described below with reference to the accompanying drawings.

FIG. 1 shows a schematic block diagram of an encoder for encoding a multi-channel audio signal. FIG. 2 shows a schematic block diagram of a linear prediction domain encoder according to the embodiment. FIG. 3 shows a schematic block diagram of a frequency domain encoder according to the embodiment. FIG. 4 shows a schematic block diagram of an audio encoder according to the embodiment. FIG. 5a shows a schematic block diagram of an active downmixer according to an embodiment. FIG. 5b shows a schematic block diagram of a passive downmixer according to an embodiment. FIG. 6 shows a schematic block diagram of a decoder for decoding an encoded audio signal. FIG. 7 shows a schematic block diagram of a decoder according to the embodiment. FIG. 8 shows a schematic block diagram of a method for encoding a multi-channel signal. FIG. 9 shows a schematic block diagram of a method for decoding an encoded audio signal. FIG. 10 shows a schematic block diagram of an encoder for encoding a multi-channel signal according to another aspect. FIG. 11 shows a schematic block diagram of a decoder for decoding an encoded audio signal according to another aspect. FIG. 12 shows a schematic block diagram of an audio encoding method for encoding a multi-channel signal according to another aspect. FIG. 13 shows a schematic block diagram of a method for decoding an encoded audio signal according to another aspect. FIG. 14 shows a schematic timing diagram of a seamless switch from frequency domain coding to LPD coding. FIG. 15 shows a schematic timing diagram of a seamless switch from frequency domain decoding to LPD domain decoding. FIG. 16 shows a schematic timing diagram of a seamless switch from LPD encoding to frequency domain encoding. FIG. 17 shows a schematic timing diagram of a seamless switch from LPD decoding to frequency domain decoding. FIG. 18 shows a schematic block diagram of an encoder for encoding a multi-channel signal according to another aspect. FIG. 19 shows a schematic block diagram of a decoder for decoding an encoded audio signal according to another aspect. FIG. 20 shows a schematic block diagram of a method of audio encoding for encoding a multi-channel signal according to another aspect. FIG. 21 shows a schematic block diagram of a method for decoding an encoded audio signal according to another aspect.

  In the following, embodiments of the present invention will be described in more detail. Elements indicated in individual numbers having the same or similar function relate to the same reference sign.

  FIG. 1 shows a schematic block diagram of an audio encoder 2 for encoding a multi-channel audio signal 4. The audio encoder includes a linear prediction domain encoder 6, a frequency domain encoder 8, and a controller 10 for switching between the linear prediction domain encoder 6 and the frequency domain encoder 8. The controller analyzes the multi-channel signal and determines whether linear prediction domain coding or frequency domain coding is advantageous for a portion of the multi-channel signal. That is, the controller is configured such that a portion of the multi-channel signal is represented by either an encoded frame of a linear prediction domain encoder or an encoded frame of a frequency domain encoder. The linear prediction domain encoder includes a downmixer 12 for downmixing the multi-channel signal 4 to obtain a downmix signal 14. The linear prediction domain encoder further includes a linear prediction domain core encoder 16 for encoding the downmix signal. Further, the linear prediction domain encoder generates first multi-channel information 20 from the multi-channel signal 4, including first multi-channel information 20 including, for example, an ILD (inter-ear level difference) parameter and / or an IPD (inter-ear phase difference) parameter. Includes a combined multi-channel encoder 18. The multi-channel signal is, for example, a stereo signal. The down mixer converts a stereo signal to a monaural signal. The linear prediction domain core encoder encodes a monaural signal. The first combined multi-channel encoder generates stereo information for the encoded monaural signal as first multi-channel information. The frequency domain encoder and controller are optional when compared to the alternative described for FIGS. 10 and 11. However, it is advantageous to use a frequency domain encoder and controller for signal adaptive switching between time domain and frequency domain coding.

  Further, the frequency domain encoder 8 includes a second combined multi-channel encoder 22 for generating second multi-channel information 24 from the multi-channel signal 4. The second combined multi-channel encoder 22 is different from the first multi-channel encoder 18. However, the second combined multi-channel processor 22 may determine that the signal better encoded by the second encoder is higher than the first reconstructed quality of the first multi-channel information obtained by the first multi-channel encoder. 2 Obtain second multi-channel information that allows re-creation quality.

  That is, according to the embodiment, the first combined multi-channel encoder 18 is configured to generate the first multi-channel information 20 that allows the first reproduction quality. The second combined multi-channel encoder 22 is configured to generate second multi-channel information 24 that allows a second reproduction quality. The second reproduction quality is higher than the first reproduction quality. This is at least relevant for signals such as, for example, speech signals. It is better encoded by the second multi-channel encoder.

  Thus, the first multi-channel encoder is a parametric combined multi-channel encoder including, for example, a stereo prediction coder, a parametric stereo encoder, or a rotation-based parametric stereo encoder. In addition, the second combined multi-channel encoder is waveform preserving, such as a band selective switch, for example for middle / side or left / right stereo coder. As described in FIG. 1, the encoded downmix signal 26 is transmitted to an audio decoder and optionally provided to a first combined multi-channel processor. For example, the coded downmix signal is a decoded signal obtained by decoding the coded signal, and the residual signal from the multi-channel signal before and after decoding is coded on the decoder side. Calculated to enhance the decoded quality of the audio signal. Further, after determining an appropriate coding scheme for the current portion of the multi-channel signal, controller 10 uses control signals 28a, 28b to control the linear prediction domain encoder and the frequency domain encoder, respectively.

  FIG. 2 shows a block diagram of the linear prediction domain encoder 6 according to the embodiment. The input to the linear prediction domain encoder 6 is a downmix signal 14 downmixed by a downmixer 12. Further, the linear prediction domain encoder includes an ACELP processor 30 and a TCX processor 32. The ACELP processor 30 is configured to operate on the downsampled downmix signal 34 that is downsampled by the downsampler 35. Further, the time domain bandwidth extension processor 36 parametrically encodes the band of the portion of the downmix signal 14 that is removed from the downsampled downmix signal 34 input into the ACELP processor 30. The time domain bandwidth extension processor 36 outputs a parametrically encoded band 38 of the portion of the downmix signal 14. That is, the time domain bandwidth extension processor 36 calculates a parametric representation of the frequency band of the downmix signal 14 that includes higher frequencies than the cutoff frequency of the downsampler 35. Thus, the down-sampler 35 may provide the time-domain bandwidth extension processor 36 with those frequency bands above the cut-off frequency of the down-sampler, or the time-domain bandwidth extension (TD-BWE) processor 36 may be down. It has another property to provide a cut-off frequency to the TD-BWE processor to enable calculating the parameter 38 for the correct part of the mix signal 14.

  Further, the TCX processor is configured to operate on the unmixed or downsampled downmix signal to a lesser extent than the downsampling for the ACELP processor, for example. Downsampling to a lesser extent than ACELP processor downsampling is downsampling using a higher cutoff frequency. Multiple bands of the downmix signal are provided to the TCX processor when compared to the downsampled downmix signal 35 input to the ACELP processor 30. The TCX processor further includes a first time-frequency converter 40, such as, for example, an MDCT, DFT or DCT. The TCX processor 32 further includes a first parameter generator 42 and a first quantizer encoder 44. A first parameter generator 42 using, for example, an intelligent gap filling (IGF) algorithm calculates a first parametric representation of a first band set 46. A first quantizer encoder 44, for example using the TCX algorithm, calculates a first set of quantized and encoded spectral lines 48 for a second set of bands. That is, the first quantizer encoder parametrically encodes an associated band, such as a tone band, of the inbound signal. The first parameter generator applies, for example, the IGF algorithm to the remaining band of the inbound signal to further reduce the bandwidth of the encoded audio signal.

  The linear predictive domain encoder 6 may, for example, quantize the ACELP downsampled downmix signal 52 and / or the first parametric representation of the first band set 46 and / or the second band set. And a linear prediction domain decoder 50 for decoding the downmix signal 14, represented by the first set of encoded spectral lines 48. The output of the linear prediction domain decoder 50 is an encoded and decoded downmix signal 54. This signal 54 is input to a multi-channel residual coder 56 that calculates and encodes a multi-channel residual signal 58 using the encoded and decoded downmix signal 54. The encoded multi-channel residual signal represents an error between the decoded multi-channel representation using the first multi-channel information and the multi-channel signal before downmixing. Accordingly, the multi-channel residual coder 56 includes the joint encoder-side multi-channel decoder 60 and the difference processor 62. The combined encoder-side multi-channel decoder 60 generates a decoded multi-channel signal using the first multi-channel information 20 and the encoded and decoded downmix signal 54. The difference processor forms a difference between the decoded multi-channel signal 64 and the multi-channel signal 4 before downmixing to obtain a multi-channel residual signal 58. That is, the combined encoder-side multi-channel decoder in the audio encoder performs the decoding operation. It is advantageously the same decoding operation performed on the decoder side. Therefore, the first combined multi-channel information derived by the audio decoder after transmission is used in the combined encoder-side multi-channel decoder for decoding the encoded downmix signal. The difference processor 62 calculates the difference between the decoded combined multi-channel signal and the original multi-channel signal 4. For example, for parametric encoding, the encoded multi-channel residual signal 58 is reduced because the difference between the decoded signal and the original signal is reduced by knowledge of the difference between these two signals. Increases the decoding quality of the audio decoder. This allows the first combined multi-channel encoder to operate in such a way that multi-channel information for the full bandwidth of the multi-channel audio signal is derived.

  Further, the downmix signal 14 includes a low band and a high band. The linear prediction domain encoder 6 is configured to apply a bandwidth extension process using, for example, a time domain bandwidth extension processor 36 for encoding a high band parametrically. The linear prediction domain decoder 6 is configured to obtain only a low-band signal representing the low band of the downmix signal 14 as the encoded and decoded downmix signal 54. The encoded multi-channel residual signal has only a frequency within a low band of the multi-channel signal before downmixing. That is, the bandwidth extension processor calculates a bandwidth extension parameter for a frequency band higher than the cutoff frequency. The ACELP processor encodes frequencies below the cutoff frequency. Accordingly, the decoder is configured to reconstruct higher frequencies based on the encoded low band signal and the bandwidth parameter 38.

  According to another embodiment, multi-channel residual coder 56 calculates a side signal. The downmix signal is a corresponding intermediate signal of the M / S multi-channel audio signal. Thus, the multi-channel residual coder computes the calculated side signal and the encoded and decoded downmix signal 54, which are calculated from the full-band spectral representation of the multi-channel audio signal obtained by the filter bank 82. The difference between the multiple predicted side signal is calculated and encoded. The multiple represented by the prediction information becomes part of the multi-channel information. However, the downmix signal includes only the low band signal. Therefore, the residual coder further calculates the residual (or side) signal for the high band. This is performed, for example, by simulating a time domain bandwidth extension, as is done in a linear prediction domain core encoder. Alternatively, it is performed by predicting the side signal as the difference between the calculated (full band) side signal and the calculated (full band) intermediate signal. The prediction factor is configured to minimize the difference between both signals.

  FIG. 3 shows a schematic block diagram of the frequency domain encoder 8 according to the embodiment. The frequency domain encoder includes a second time-frequency converter 66, a second parameter generator 68, and a second quantizer encoder 70. The second time-frequency converter 66 converts the first channel 4a and the second channel 4b of the multi-channel signal into spectral representations 72a, 72b. The first channel spectral representation 72a and the second channel spectral representation 72b are analyzed and split into a first band set 74 and a second band set 76, respectively. Accordingly, the second parameter generator 68 generates a second parametric representation 78 of the second band set 76. The second quantizer encoder produces a quantized encoded representation 80 of the first set of bands 74. The frequency domain encoder, more specifically, the second time-frequency converter 66 performs, for example, an MDCT operation on the first channel 4a and the second channel 4b. The second parameter generator 68 executes an intelligent gap filling algorithm, and the second quantizer encoder 70 performs, for example, an AAC operation. Thus, as already described for the linear prediction domain encoder, the frequency domain encoder is operable in such a way that multi-channel information for the entire bandwidth of the multi-channel audio signal is derived.

  FIG. 4 shows a schematic block diagram of the audio encoder 2 according to the preferred embodiment. The LPD path 16 consists of a combined stereo or multi-channel encoding that includes an “active or passive DMX” downmix calculation 12, as shown in FIG. Target)) or passive ("constant mixed factor"). The downmix is further encoded by a switchable mono ACELP / TCX core supported by either the TD-BWE or IGF module. Note that ACELP operates on downsampled input audio data 34. Any ACELP initialization by switching is performed on the downsampled TCX / IGF output.

  Since ACELP does not include any internal time-frequency decomposition, LPD stereo coding is particularly complicated by the means of an analysis filter bank 82 before LP coding and a synthesis filter bank after LPD decoding. Add a filter bank for any module. In a preferred embodiment, an oversampled DFT with low overlap is employed. However, in another embodiment, an oversampled time-frequency decomposition with similar temporal resolution can be used. The stereo parameters are then calculated in the frequency domain.

  Parametric stereo encoding is performed by an “LPD stereo parameter encoding” block 18 that outputs LPD stereo parameters 20 to a bitstream. Optionally, the following block "LPD Stereo Residual Encoding" adds the vector quantized low pass downmix residual 58 to the bitstream.

  FD path 8 is configured to have a combined stereo or multi-channel encoding within itself. For joint stereo coding, it reuses its own critically sampled and real-valued filterbank 66, eg, the MDCT.

  The signals provided to the decoder are multiplexed, for example, into a single bit stream. The bitstream comprises an encoded downmix signal 26 further including at least one of the parametrically encoded time domain bandwidth extended bands 38, an ACELP processed downsampled downmix signal 52, , The first multi-channel information 20, the coded multi-channel residual signal 58, the first parametric representation of the first band set 46, and the quantized and coded spectrum for the second band set 48. A first set of lines and second multi-channel information 24 including a quantized encoded representation of a first set of bands 80 and a second parametric representation of a first set 78 of bands.

  Embodiments provide switchable core encoders, combined multi-channel coding and parametric spatial audio coding, completely depending on the choice of core encoder, allowing different multi-channel coding techniques to be used. 2 shows an improved method for coupling to a switchable perceptual encoder. In particular, within a switchable audio encoder, native frequency domain stereo coding is combined with ACELP / TCX based on linear predictive coding, which has its own dedicated independent parameter stereo coding.

  5a and 5b show an active and a passive downmixer, respectively, according to an embodiment. The active downmixer operates in the frequency domain, for example, using a time frequency converter 82 for converting the time domain signal 4 into a frequency domain signal. After downmixing, a frequency-to-time transform (e.g., IDFT) converts the signal downmixed from the frequency domain into a downmixed signal 14 in the time domain.

  FIG. 5b shows a passive downmixer 12 according to an embodiment. The passive downmixer 12 includes an adder to which the first channel 4a and the second channel 4b are respectively combined after being weighted using the weights 84a and 84b. Further, the first channel 4a and the second channel 4b are input to the time-frequency converter 82 before transmission to LPD stereo parametric coding.

  That is, the downmixer is configured to convert the multi-channel signal into a spectral representation. Downmixing is performed using a spectral representation or using a time domain representation. The first multi-channel encoder is configured to use the spectral representation to generate separate first multi-channel information for each band of the spectral representation.

  FIG. 6 shows a schematic block diagram of an audio decoder 102 for decoding an encoded audio signal 103 according to the embodiment. The audio decoder 102 includes a linear prediction domain decoder 104, a frequency domain decoder 106, a first combined multi-channel decoder 108, a second multi-channel decoder 110, and a first combiner 112. The encoded audio signal 103, which is a multiplexed bitstream of the previously described encoder portion, such as a frame of an audio signal, for example, by a combined multi-channel decoder 108 using the first multi-channel information 20, or The multi-channel is decoded by the frequency domain decoder 106 and the multi-channel decoded by the second combined multi-channel decoder 110 using the second multi-channel information 24. The first combined multi-channel decoder outputs a first multi-channel representation 114, and the output of the second combined multi-channel decoder 110 is a second multi-channel representation 116.

  That is, the first combined multi-channel decoder 108 generates the first multi-channel representation 114 using the output of the linear prediction domain encoder and the first multi-channel information 20. The second multi-channel decoder 110 generates a second multi-channel representation 116 using the output of the frequency domain decoder and the second multi-channel information 24. Further, the first combiner combines the first multi-channel representation 114 and the second multi-channel representation 116 based on, for example, the frame to obtain a decoded audio signal 118. Further, the first combined multi-channel decoder 108 is, for example, a parametric combined multi-channel decoder that uses complex prediction, parametric stereo operations or rotation operations. The second combined multi-channel decoder 110 is, for example, a waveform-maintaining combined multi-channel decoder that uses band-selective switches for middle / side or left / right stereo decoding algorithms.

  FIG. 7 shows a schematic block diagram of a decoder 102 according to another embodiment. Here, the linear prediction domain decoder 102 includes an ACELP decoder 120, a low-band synthesizer 122, an up-sampler 124, a time-domain bandwidth extension processor 126, or a second signal for combining the up-sampled signal and the bandwidth extension signal. Includes a two combiner 128. In addition, the linear prediction domain decoder includes a TCX decoder 132 and an intelligent gap filling processor 132, described as one block in FIG. Further, the linear prediction domain decoder 102 includes a full-band synthesis processor 134 for combining the outputs of the second combiner 128, the TCX decoder 130, and the IGF processor 132. As already shown for the encoder, the time domain bandwidth extension processor 126, ACELP decoder 120 and TCX decoder 130 work in parallel to decode the individual transmitted audio information.

  A cross-path 136 is provided for initializing the low-band synthesizer using information derived from the low-band spectral-time transform using, for example, a frequency-to-time converter 138 from the TCX decoder 130 and the IGF processor 132. By referencing the vocal spread model, the ACELP data models the vocal spread. The TCX data models the excitation of the vocal spread. For example, the cross-path 136 represented by a low-band frequency-to-time converter, such as an IMDCT decoder, indicates that the low-band synthesizer 122 uses a vocal-spread form, and that the current excitation is a coded low-band signal. Can be recalculated or decoded. Further, the synthesized low band is up-sampled by the up-sampler 124 and the second sampled to recreate the up-sampled frequency, e.g., to recover energy for each up-sampled band, e.g. It is combined with the time domain bandwidth extension high band 140 using a combiner 128.

  The full band synthesizer 134 uses the full band signal of the second combiner 128 and the excitation from the TCX processor 130 to form a decoded downmix signal 142. The first combined multi-channel decoder 108 includes a time-frequency converter 144 for converting the output of the linear prediction domain decoder, for example, the decoded downmix signal 142 into a spectral representation 145. Further, an upmixer implemented, for example, in the stereo decoder 146 is controlled by the first multi-channel information 20 to upmix the spectral representation to a multi-channel signal. Further, frequency-time-converter 148 converts the result of the upmix into time representation 114. Time-frequency and / or frequency-time-converters include complex operations such as DFT or IDFT or oversampled operations.

  Further, a first combined multi-channel decoder, or more specifically, a stereo decoder 146, provides a multi-channel residual signal provided by, for example, a multi-channel encoded audio signal 103 to generate a first multi-channel representation. Use 58. Further, the multi-channel residual signal includes a lower bandwidth than the first multi-channel representation. The first combined multi-channel decoder reconstructs the intermediate first multi-channel representation using the first multi-channel information and adds the multi-channel residual signal to the intermediate first multi-channel representation. Be composed. That is, the stereo decoder 146 performs multi-channel decoding using the first multi-channel information 20 after the decoded spectral representation of the down-mix signal is up-mixed into the multi-channel signal, and optionally multi-channel decoding. Improving the reconstructed multi-channel signal by adding the channel residual signal to the reconstructed multi-channel signal. Therefore, the first multi-channel information and the residual signal already act on the multi-channel signal.

  The second combined multi-channel decoder 110 uses as input the spectral representation obtained by the frequency domain decoder. The spectral representation includes a first channel signal 150a and a second channel signal 150b for at least a plurality of bands. Further, the second combined multi-channel processor 110 adapts to a plurality of bands of the first channel signal 150a and the second channel signal 150b. For example, a combined multi-channel operation, such as a mask, displays a left / right or middle / side combined multi-channel encoding for each band. A combined multi-channel operation is an intermediate / side or left / right conversion operation to convert a band displayed from a middle / side representation to a left / right representation by a mask. It transforms the result of the combined multi-channel operation into a temporal representation to obtain a second multi-channel representation. Furthermore, the frequency domain decoder includes a frequency-to-time converter 152, for example an IMDCT operation or especially a sampled operation. That is, the mask includes a flag indicating, for example, L / R or M / S stereo encoding. The second combined multi-channel encoder applies a corresponding stereo coding algorithm to individual audio frames. Optionally, intelligent gap filling is applied to the encoded audio signal to further reduce the bandwidth of the encoded audio signal. Thus, for example, the tone frequency band is encoded at high resolution using the stereo encoding algorithm described above. Other frequency bands are coded parametrically, for example by using the IGF algorithm.

  That is, in the LPD path 104, the transmitted monaural signal is reconstructed by a switchable ACELP / TCX 120/130 decoder supported, for example, by a TD-BWE 126 or IGF module 132. Any ACELP initialization by switching is performed on the downsampled TCX / IGF output. The output of the ACELP is upsampled to a full sampling rate using, for example, an upsampler 124. All signals are mixed in the time domain at a high sampling rate using, for example, mixer 128 and further processed by LPD stereo decoder 146 to provide LPD stereo.

  LPD “stereo decoding” consists of an upmix of the transmitted downmix, guided by the application of the transmitted stereo parameters 20. Optionally, a downmix residual 58 is included in the bitstream. In this case, the residual is decoded and included in the upmix calculation by "stereo decoding" 146.

  The FD path 106 is configured to have its own independent internal combined stereo or multi-channel decoding. For combined stereo, decoding it reuses a critically sampled, real-valued filterbank 152, eg, IMDCT.

  The LPD stereo output and the FD stereo output are mixed in the time domain, for example using the first combiner 112, to provide a final output 118 of the fully switched encoder.

  Even though multi-channels are described for stereo decoding in the relevant figures, the same principles also apply to multi-channel processing, generally with more than one channel.

  FIG. 8 shows a schematic block diagram of a method 800 for encoding a multi-channel signal. The method 800 includes performing 805 linear prediction domain coding, performing 810 frequency domain coding, and switching 815 between linear prediction domain coding and frequency domain coding. The step of performing linear prediction domain coding includes a downmix signal, a linear prediction domain that core codes the downmix signal, and a first combined multichannel coding that generates first multichannel information from a multichannel signal. Down-mixing the multi-channel signal to obtain. Frequency domain encoding includes performing a second combined multi-channel encoding that generates second multi-channel information from the multi-channel signal. The step of encoding the second combined multi-channel is different from the step of encoding the first multi-channel. The switching is performed such that a portion of the multi-channel signal is represented by either a coded frame of linear prediction domain coding or a coded frame of frequency domain coding.

  FIG. 9 shows a schematic block diagram of a method 900 for decoding an encoded audio signal. The method 900 comprises a step 905 of linear prediction domain decoding, a step 910 of frequency domain decoding, and a first combination for generating a first multi-channel representation using an output of the linear prediction domain decoding and the first multi-channel information. Multi-channel decoding 915; second multi-channel decoding 920 for generating a second multi-channel representation using the output of the frequency domain decoding and the second multi-channel information; Combining 925 the first multi-channel representation and the second multi-channel representation to obtain. The step of decoding the second first multi-channel information is different from the step of decoding the first multi-channel information.

  FIG. 10 shows a schematic block diagram of an audio encoder for encoding a multi-channel signal according to another aspect. The audio encoder 2 'includes a linear prediction domain encoder 6 and a multi-channel residual encoder 56. The linear prediction domain encoder includes a downmixer 12 for downmixing the multi-channel signal 4 and a linear prediction domain core encoder 16 for encoding the downmix signal 14 to obtain the downmix signal 14. Including. The linear prediction domain encoder 6 further includes a combined multi-channel encoder 18 for generating multi-channel information 20 from the multi-channel signal 4. Further, the linear prediction domain encoder includes a linear prediction domain decoder 50 for decoding the encoded downmix signal 26 to obtain an encoded and decoded downmix signal 54. The multi-channel residual encoder 56 calculates and encodes a multi-channel residual signal using the coded and decoded downmix signal 54. The multi-channel residual signal represents an error between the decoded multi-channel representation 54 using the multi-channel information 20 and the multi-channel signal 4 before downmixing.

  According to the embodiment, the downmix signal 14 includes a low band and a high band. The linear prediction domain decoder uses a bandwidth extension processor to apply bandwidth extension processing to parametric encoding of high bands. The linear prediction domain encoder is configured to obtain only the low-band signal representing the low band of the downmix signal as the encoded and decoded downmix signal 54. The encoded multi-channel residual signal has only a band corresponding to the low band of the multi-channel signal before downmixing. Furthermore, the same description for audio encoder 2 applies to audio encoder 2 '. However, another frequency encoding of the encoder 2 is omitted. This simplifies the encoder configuration and is therefore advantageous if the encoder is used only for audio signals containing signals. It is encoded parametrically in the time domain with no noticeable quality loss or the quality of the decoded audio signal is still within the standard. However, dedicated residual stereo coding is advantageous for increasing the reproduction quality of the decoded audio signal. More specifically, the difference between the encoded audio signal and the decoded audio signal is known by the decoder, so the difference between the audio signal before encoding and the encoded and decoded audio signal is The difference is derived and sent to a decoder to increase the reproduction quality of the decoded audio signal.

  FIG. 11 shows an audio decoder 102 'for decoding an encoded audio signal 103 according to another aspect. The audio decoder 102 'includes a linear prediction domain decoder 104 and a combined multi-channel decoder 108 for generating a multi-channel representation 114 using the output of the linear prediction domain decoder 104 and the combined multi-channel information 20. Further, the encoded audio signal 103 includes a multi-channel residual signal 58 used by a multi-channel decoder to generate a multi-channel representation 114. Further, the same description in relation to audio decoder 102 applies to audio decoder 102 '. Here, the residual signal from the original audio signal to the decoded audio signal is compared with the original audio signal, even if parametric and therefore wasteful coding is used. Used and applied to the decoded audio signal to achieve at least almost the same quality of the decoded audio signal. However, the frequency decoding portion shown for audio decoder 102 is omitted in audio decoder 102 '.

  FIG. 12 shows a schematic block diagram of an audio encoding method 1200 for encoding a multi-channel signal. The method 1200 includes a step 1205 of performing linear prediction domain coding including downmixing of the multi-channel signal to obtain a down-mixed multi-channel signal. The linear prediction domain core encoder generates multi-channel information from the multi-channel signal. The method further includes a linear prediction domain for decoding the downmix signal to obtain an encoded and decoded downmix signal. The method 1200 includes a multi-channel residual encoding step 1210 of calculating an encoded multi-channel residual signal using the encoded and decoded downmix signal. The multi-channel residual signal represents an error between the decoded multi-channel representation using the first multi-channel information and the multi-channel signal before downmixing.

  FIG. 13 shows a schematic block diagram of a method 1300 for decoding an encoded audio signal. The method 1300 includes linear predictive domain decoding 1305 and combined multi-channel decoding 1310 using the output of the linear predictive domain decoding and the combined multi-channel information to generate a multi-channel representation. The encoded multi-channel audio signal includes a channel residual signal. Joint multi-channel decoding uses the multi-channel residual signal to generate a multi-channel representation.

  The described embodiments cover all types of stereo or multi-channel audio content (similar speech and music with a constant perceived quality at a given low bit rate), for example digital radio, internet streaming and audio communication applications For use in distribution of broadcasts.

  14 to 17 illustrate embodiments of how to apply the proposed seamless switching between LPD coding and frequency domain coding. The reverse is also true. In general, past windowing or processing is indicated using thin lines, and thicker lines indicate current windowing or processing. Switching is applied, and the dotted line indicates the current processing made exclusively for the transition or switching. Switching or transition from LPD encoding to frequency encoding.

  FIG. 14 shows a schematic timing diagram displaying an embodiment for seamless switching between frequency domain coding and time domain coding. This is appropriate if, for example, the controller 10 indicates that the current frame is better encoded using LPD encoding instead of the FD encoding used for the previous frame. During frequency domain coding, stop windows 200a and 200b are applied to each stereo signal (optionally extended to two or more channels). The stop window differs from the standard MDCT convolution add fade at the beginning 202 of the first frame 204. The left side of the stop window is a traditional convolution sum to encode the previous frame, for example, using an MDCT time-frequency transform. Therefore, the frame before the switch is still properly encoded. For the current frame 204, if the switch is applied, even if the first parametric representation of the intermediate signal for time domain encoding is calculated for the following frame 206, the additional stereo parameters will be Is calculated. These two additional stereo analyzes are made so that an intermediate signal 208 can be generated for the LPD look ahead. However, the stereo parameters are transmitted (in addition) for the two first LPD stereo windows. In the normal case, the stereo parameters are sent with two LPD stereo frames with delay. The intermediate signal is also used for the past to update the ACELP memory, such as LPC analysis or Forward Alias Cancellation (FAC). Later, the LPD stereo windows 210a-210d for the first stereo signal and the LPD stereo windows 212a-212d for the second stereo signal are analyzed before applying a time-frequency transform using, for example, DFT. Applied in bank 82. The intermediate signal includes a typical crossfade slope when using TCX coding, resulting in an exemplary LPD analysis window 214. If ACELP is used to encode an audio signal, such as a mono low-band signal, it simply selects a plurality of frequency bands indicated by a rectangular LPD analysis window 216 to which LPC analysis is applied.

  Further, the timing indicated by the vertical line 218 indicates that the current frame to which the transition is applied includes information from the frequency domain analysis windows 200a, 200b and the calculated intermediate signal 208 and the corresponding stereo information. During the horizontal portion of the frequency analysis window between line 202 and line 218, frame 204 is fully encoded using frequency domain coding. From line 218 to the end of the frequency analysis window on line 220, frame 204 contains information from both frequency domain and LPD encoding, and from line 220 to the end of frame 204 on vertical line 222, the LPD code Only the encoding contributes to the encoding of the frame. Even more attention is drawn in the middle of the coding, as the first and last (third) parts are easily derived from one coding technique without aliasing. However, for the middle part, it should be distinguished between ACELP and TCX mono signal encoding. TCX coding uses crossfades, as already applied by frequency domain coding, so that simple fades outside the frequency coded signal and fade-in of the TCX coded intermediate signal , Provides complete information for encoding the current frame 204. If ACELP is used for monaural signal encoding, more sophisticated processing is applied since area 224 does not contain complete information for encoding the audio signal. The proposed method is, for example, Forward Aliasing Correction (FAC) as described in the USAC standard in section 7.16.

  According to an embodiment, the controller 10 uses the frequency domain encoder 8 to encode the previous frame within the current frame 204 of the multi-channel audio signal, and thus the linear controller to decode the subsequent frame. It is configured to switch to the prediction domain encoder. First combined multi-channel encoder 18 calculates composite multi-channel parameters 210a, 210b, 212a, 212b from the multi-channel audio signal for the current frame. The second combined multi-channel encoder 22 is configured to weight the second multi-channel signal using a stop window.

FIG. 15 shows a schematic timing diagram of a decoder corresponding to the encoder operation of FIG. Here, the reconstruction of the current frame 204 will be described according to the embodiment. As already shown in the encoder timing diagram of FIG. 14, the frequency domain stereo channel is provided from the frame before applying the stop windows 200a and 200b. The transition from FD to LPD mode is made first in the decoded intermediate signal, as in the mono case. It is achieved by artificially creating an intermediate signal 226 from the decoded time domain signal 116 in FD mode. ccfl is the core code frame length, and L_fac indicates the length of the frequency aliasing cancellation window or frame or block or transform.

  This signal is then passed to the LPD decoder 120 for updating the memory and applying the FAC to decode, as is done in the case of a monaural transition from FD mode to ACELP. The processing is described in section 7.16 of the USAC standard [ISO / IEC DIS 23003-3, Usac]. In the case of switching from the FD mode to the TCX, the conventional superposition and addition is performed. The LPD stereo decoder 146 applies the time-frequency conversion of the time-frequency converter 144 as an input signal to the already transposed stereo processing, for example, by applying the transmitted stereo parameters 210 and 212. Receive the decoded intermediate signal (in the later frequency domain). The stereo decoder then outputs left and right channel signals 228, 230 that overlap the previous frame decoded in FD mode. The signals, i.e., the FD decoded time domain signal and the LPD decoded time domain signal for the frame to which the transition is applied, are then split into individual channels (combiner 112) to smooth the transition in the left and right channels. Cross-fade).

  In FIG. 15, the transition is illustrated schematically using M = ccfl / 2. In addition, the combiner performs crossfading on successive frames being decoded, using only FD or LPD decoding, without transitions between these modes.

  That is, the superposition addition processing of the FD decoding is replaced by the cross-fade of the FD decoded audio signal and the LPD decoded audio signal, particularly when using the MDCT / IMDCT for time-frequency / frequency-time conversion. Therefore, the decoder should calculate the LPD signal for the fade-out portion of the FD decoded audio signal in order to fade in the LPD decoded audio signal. According to an embodiment, the audio decoder 102 uses the frequency domain decoder 106 to decode the previous frame within the current frame 204 of the multi-channel audio signal, and thus uses the frequency domain decoder 106 to decode the subsequent frame. It is configured to switch to the linear prediction domain decoder 104. The combiner 112 calculates a composite intermediate signal 226 from the second multi-channel representation 116 of the current frame. First combined multi-channel decoder 108 generates first multi-channel representation 114 using composite intermediate signal 226 and first multi-channel information 20. Further, the combiner 112 is configured to combine the first multi-channel representation and the second multi-channel representation to obtain a decoded current frame of the multi-channel audio signal.

  FIG. 16 shows a schematic timing diagram in an encoder for performing a transition from using LPD encoding to using FD decoding within a current frame 232. To switch from LPD encoding to FD encoding, start windows 300a, 300b are applied to FD multi-channel encoding. The start window has a similar function when compared to the stop windows 200a, 200b. During the fade-out of the TCX encoded mono signal of the LPD encoder between vertical lines 234 and 236, start windows 300a, 300b perform a fade-in. When using ACELP instead of TCX, monaural signals do not perform a smooth fade-out. Nevertheless, the correct audio signal is reconstructed at the decoder using, for example, FAC. LPD stereo windows 238 and 240 are calculated by default and refer to the ACELP or TCX coded monaural signal, indicated by LPD analysis window 241.

  FIG. 17 shows a schematic timing diagram in a decoder corresponding to the timing diagram of the encoder described for FIG.

  For the transition from LPD mode to FD mode, special frames are decoded by stereo decoder 146. The intermediate signal coming from the LPD mode decoder is extended with zero for the frame index i = ccfl / M.

  The previously described stereo decoding is performed by retaining the last stereo parameters, and switching off sets the side signal dequantization, ie, code_mode to zero. Furthermore, right windowing after the inverse DFT is not applied, which results in sharp edges 242a, 242b of special LPD stereo windows 244a, 244b. It can be clearly seen that the specific shaped edges are located in the flat sections 246a, 246b. The entire information of the corresponding part of the frame is derived from the FD encoded audio signal. Thus, right-hand windowing (without sharp edges) results in unwanted interference from LPD information to FD information and is therefore not applied.

  The resulting left and right (decoded LPD) channels 250a, 250b (using the LPD analysis window 248 and the LPD decoded intermediate signal indicated by the stereo parameters) are then processed when in TCX to FD mode. By using superimposed summation or by using FAC on a per-channel basis when going from ACELP to FD mode. A summary description of the transition is set forth in FIG. Here, M = ccfl / 2.

  According to an embodiment, the audio decoder 102 uses the linear prediction domain decoder 104 to decode the previous frame within the current frame 232 of the multi-channel audio signal, and thus decodes the subsequent frame. To the frequency domain decoder 106. Stereo decoder 146 uses the multi-channel information of the previous frame to calculate a composite multi-channel audio signal from the linear prediction domain decoder decoded monaural signal for the current frame. The second combined multi-channel decoder 110 calculates a second multi-channel representation for the current frame and weights the second multi-channel representation using a start window. A combiner 112 combines the combined multi-channel audio signal and the weighted second multi-channel representation to obtain a decoded current frame of the multi-channel audio signal.

  FIG. 18 shows a schematic block diagram of an encoder 2 ″ for encoding the multi-channel signal 4. The audio encoder 2 ″ includes a down mixer 12, a linear prediction domain core encoder 16, a filter bank 82, and a combined multi-channel encoder 18. The downmixer 12 is configured to downmix the multi-channel signal 4 to obtain a downmix signal 14. The downmix signal is a monaural signal such as an intermediate signal of an M / S multi-channel audio signal. The linear prediction domain core encoder 16 encodes the downmix signal 14. The downmix signal 14 has a low band and a high band. The linear prediction domain core encoder 16 is configured to apply bandwidth extension processing and apply coding in a high band parametrically. Further, the filter bank 82 generates a spectral representation of the multi-channel signal 4. Combined multi-channel encoder 18 is configured to process a spectral representation of the multi-channel signal including the low and high bands to generate multi-channel information 20. The multi-channel information includes ILD and / or IPD and / or IID (mutual hearing intensity difference) parameters that allow the decoder to recalculate the multi-channel audio signal from the monaural signal. A more detailed view of another aspect of an embodiment according to this aspect is found in the previous figure, particularly FIG.

  According to the embodiment, the linear prediction domain core encoder 16 further includes a linear prediction domain decoder for decoding the encoded downmix signal 26 and obtaining an encoded and decoded downmix signal 54. Including. Here, the linear prediction domain core encoder forms an intermediate signal of the M / S audio signal that is encoded for transmission to the decoder. Further, the audio encoder further includes a multi-channel residual encoder 56 for calculating an encoded multi-channel residual signal 58 using the encoded and decoded downmix signal 54. The multi-channel residual signal uses the multi-channel information 20 to represent the error between the decoded multi-channel representation and the multi-channel signal 4 before downmixing. That is, the multi-channel residual signal 58 is a side signal of the M / S audio signal and corresponds to the intermediate signal calculated using the linear prediction domain core encoder.

  According to another embodiment, the linear prediction domain core encoder 16 applies a bandwidth extension process to parametrically encode the high band, and outputs the encoded and decoded downmix signal as a downmix signal. It is configured to obtain only the low band signal representing the low band of the mix signal. The encoded multi-channel residual signal 58 has only a band corresponding to the low band of the multi-channel signal before downmixing. Additionally or alternatively, the multi-channel residual encoder simulates a time-domain bandwidth extension applied to the high-band of the multi-channel signal in a linear prediction domain core encoder to obtain a residual for the high-band. Or, calculate the side signal to enable more accurate decoding of the monaural or intermediate signal to derive a decoded multi-channel audio signal. The simulation includes the same or similar calculations performed in the decoder to decode the bandwidth extension high band. An alternative or additional approach for simulating bandwidth extension is side signal prediction. Therefore, the multi-channel residual encoder calculates the full-band residual signal from the parametric representation 83 of the multi-channel audio signal 4 after the time-frequency conversion in the filter bank 82. This full band side signal is compared with the frequency representation of the full band intermediate signal, which is also derived from the parameter representation 83. The full band intermediate signal is calculated, for example, as the sum of the left and right channels of the parametric representation 83, and the full band side signal is calculated as the difference therefrom. Thus, the prediction further calculates the prediction factor of the full band intermediate signal that minimizes the absolute difference of the full band side signal, and the creation of the prediction factor and the full band intermediate signal.

  That is, the linear prediction domain encoder is configured to calculate the downmix signal 14 as a parametric representation of an intermediate signal of the M / S multi-channel audio signal. The multi-channel residual encoder is configured to calculate a side signal corresponding to an intermediate signal of the M / S multi-channel audio signal. The residual encoder uses the simulation time domain bandwidth extension to calculate the high band of the intermediate signal. Or, the residual encoder predicts the high band of the intermediate signal using discovery of prediction information that minimizes the difference between the side signal calculated from the previous frame and the calculated full-band intermediate signal. .

  Another embodiment shows a linear prediction domain core encoder 16 that includes an ACELP processor 30. The ACELP processor operates on the downsampled downmix signal 34. Further, the time domain bandwidth extension processor 36 is configured to parametrically encode the bandwidth of the portion of the downmix signal that has been removed from the ACELP input signal by the third downsampling. Additionally or alternatively, linear prediction domain core encoder 16 includes a TCX processor 32. The TCX processor 32 operates on the downmixed signal 14 which is not downsampled or to a lesser extent than the downsampling for the ACELP processor. Further, the TCX processor includes a first time-frequency converter 40, a first parameter generator 42 for generating a parametric representation 46 of the first band set, and a quantized and encoded for the second band set. And a first quantizer encoder 44 for generating a set of spectral lines 48. The ACELP processor and the TCX processor may, for example, be that the first number of frames is encoded using ACELP and the second number of frames is encoded using TCX, or both ACELP and TCX , Contributing information to decode one frame in a combining method.

Another embodiment shows the time-frequency converter 40 different from the filter bank 82. Filter bank 82 includes filter parameters optimized to generate a spectral representation 83 of multi-channel signal 4. Time-frequency converter 40 includes filter parameters optimized to generate a parametric representation 46 of the first set of bands. Note that in another step, the linear prediction domain encoder uses a different filter bank, or even no filter bank, for bandwidth extension and / or ACELP. Further, the filter bank 82 computes separate filter parameters to generate the spectral representation 83 without relying on previous parameter selections of the linear prediction domain encoder. That is, multi-channel coding in LPD mode uses a filter bank for multi-channel processing (DFT) that is not used in bandwidth extension (time domain for ACELP and MDCT for TCX). The advantage is that individual parametric coding can use its optimal time-frequency decomposition to obtain its parameters. For example, a combination of ACELP + TDBWE with parametric multi-channel coding with an external filter bank (eg, DFT) is advantageous. This combination is particularly efficient because the best bandwidth extension for speech is in the time domain and multi-channel processing is known to be in the frequency domain. Since ACELP + TDBWE does not have any time-frequency converters, an external filter bank or transform such as DFT is preferred or even necessary. Other concepts always use the same filter bank and therefore do not use different filter banks, for example:
-IGF and combined stereo coding for AAC of MDCT-SBR + PS for HeAACv2 of QMF
SBR + MPS212 for QMF USAC.

  According to another embodiment, the multi-channel encoder includes a first frame generator, and the linear prediction domain core encoder includes a second frame generator. The first and second frame generators are configured to form a frame from the multi-channel signal 4. The first and second frame generators are configured to form frames of equal length. That is, the framing of the multi-channel processor is the same as that used in ACELP. Even if multi-channel processing is done in the frequency domain, the time resolution for calculating its parameters or downmix is close to, or even equal to, ACELP framing. The equivalent length in this case relates to ACELP framing which is equal to or close to the time resolution for calculating the parameters for multi-channel processing or downmixing.

  According to another embodiment, the audio encoder comprises a linear prediction domain encoder 6, including a linear prediction domain core encoder 16 and a multi-channel encoder 18, a frequency domain encoder 8, and a linear prediction domain encoder 6 and a frequency domain encoder 8. And a controller 10 for switching between the two. Frequency domain encoder 8 includes a second combined multi-channel encoder 22 for encoding second multi-channel information 24 from the multi-channel signal. The second combined multi-channel encoder 22 is different from the first combined multi-channel encoder 18. Further, controller 10 is configured such that a portion of the multi-channel signal is represented by either a coded frame of a linear prediction domain encoder or a coded frame of a frequency domain encoder.

  FIG. 19 is a schematic block diagram of a decoder 102 ″ for decoding an encoded audio signal 103 including a core encoded signal, a bandwidth extension parameter, and multi-channel information according to another aspect. Is shown. The audio decoder includes a linear prediction domain core decoder 104, an analysis filter bank 144, a multi-channel decoder 146, and a synthesis filter bank processor 148. The linear prediction domain core decoder 104 decodes the core-encoded signal to generate a monaural signal. This is the (full band) intermediate signal of the M / S encoded audio signal. The analysis filter bank 144 converts the monaural signal into a spectral representation 145. The multi-channel decoder 146 generates a first channel spectrum and a second channel spectrum from the monaural signal spectral representation and the multi-channel information 20. Therefore, the multi-channel decoder uses, for example, multi-channel information including a side signal corresponding to the decoded intermediate signal. A synthesis filter bank processor 148 is configured for synthesizing the first channel spectrum to obtain a first channel signal and for synthesizing the second channel spectrum to obtain a second channel signal. . Thus, preferably, the reverse operation as compared to the analysis filterbank 144 is applied to the first and second channel signals that are IDFT if the analysis filterbank uses DFT. However, a filter bank processor processes the two channel spectra using, for example, the same filter bank, for example, in parallel or in sequential order. A more detailed drawing of this alternative embodiment can be seen with respect to the previous drawing, particularly with reference to FIG.

  According to another embodiment, the linear prediction domain core decoder generates a high-band portion 140 from the bandwidth extension parameters and the low-band mono signal or the core-coded signal to generate a decoded high-band portion of the audio signal. A bandwidth extension processor 126 for obtaining 140 is included. The low band signal processor is configured to decode the low band mono signal. The combiner 128 is configured to calculate a full band mono signal using the decoded low band mono signal of the audio signal and the decoded high band of the audio signal. A low-band monaural signal is, for example, a baseband representation of an intermediate signal of an M / S multi-channel audio signal. The bandwidth extension parameters are applied to calculate a full band mono signal (in combiner 128) from the low band mono signal.

  According to another embodiment, the linear prediction domain decoder includes an ACELP decoder 120, a low band synthesizer 122, an upsampler 124, a time domain bandwidth extension processor 126, or a second combiner 128. The second combiner 128 is configured to combine the upsampled low-band signal and the bandwidth-extended high-band signal 140 to obtain a full-band ACELP-decoded monaural signal. The linear prediction domain decoder further includes a TCX decoder 130 and an intelligent gap filling processor 132 to obtain a full band TCX decoded monaural signal. Therefore, full-band synthesis processor 134 combines the full-band ACELP-decoded monaural signal with the full-band TCX-decoded monaural signal. Further, a cross-path 136 is provided for initializing the low-band synthesizer using information derived from the TCX decoder and the IGF processor by low-band spectral time transform.

  According to another embodiment, the audio decoder uses the frequency domain decoder 106 and the output 22 of the frequency domain decoder 106 and the second multi-channel information 24 to generate a second combination for generating a second multi-channel representation 116. It includes a multi-channel decoder 110 and a first combiner 112 for combining the first and second channel signals into a second multi-channel representation 116 to obtain a decoded audio signal 118. The second combined multi-channel decoder is different from the first combined multi-channel decoder. Thus, the audio decoder uses LPD or frequency domain decoding to switch between parametric multi-channel decoding. This approach has already been described in detail in the previous figures.

  According to another embodiment, the analysis filter bank 144 includes a DFT to transform the monaural signal into a spectral representation 145. The full-band synthesis processor 148 includes an IDFT for converting the spectral representation 145 into first and second channel signals. In addition, the analysis filterbank sets the window so that the previous frame and the current frame are continuous and the right part of the previous frame's spectral representation and the left part of the current frame's spectral representation overlap. Apply to the DFT-transformed spectral representation 145. That is, crossfading is applied from one DFT block to another to perform a smooth transition between successive DFT blocks and / or reduce blocking artifacts.

  According to another embodiment, multi-channel decoder 146 is configured to derive first and second channel signals from monaural signals. A monaural signal is an intermediate signal of a multi-channel signal. The multi-channel decoder 146 is configured to obtain an M / S multi-channel decoded audio signal. The multi-channel decoder is configured to calculate a side signal from the multi-channel information. Further, the multi-channel decoder 146 is configured to calculate an L / R multi-channel decoded audio signal from the M / S multi-channel decoded audio signal. The multi-channel decoder 146 calculates an L / R multi-channel decoded audio signal for a low band using the multi-channel information and the side signal. Additionally or alternatively, the multi-channel decoder 146 calculates a predicted side signal from the intermediate signal. The multi-channel decoder is further configured to calculate an L / R multi-channel decoded audio signal for high band using the predicted side signal and an ILD value of the multi-channel information.

  Further, the multi-channel decoder 146 is further configured to perform complex operations on the L / R decoded multi-channel audio signal. The multi-channel decoder uses the energy of the encoded intermediate signal and the energy of the decoded L / R multi-channel audio signal to calculate the magnitude of the complex operation to obtain energy compensation. Further, the multi-channel decoder is configured to calculate the phase of the complex operation using the IPD value of the multi-channel information. After decoding, the energy, level or phase of the decoded multi-channel signal is different from the decoded mono signal. Thus, complex operations are determined such that the energy, level or phase of the multi-channel signal matches the value of the decoded mono signal. Furthermore, the phase is adapted to the value of the phase of the multi-channel signal before encoding, for example, using IPD parameters calculated from the multi-channel information calculated on the encoder side. Further, the human perception of the decoded multi-channel signal adapts to the human perception of the original multi-channel signal prior to encoding.

  FIG. 20 shows a schematic description of a flowchart of a method 2000 for encoding a multi-channel signal. The method includes a step 2050 of downmixing a multi-channel signal to obtain a downmix signal, and a step 2100 of encoding the downmix signal. The downmix signal has a low band and a high band. The linear prediction domain core encoder is configured to apply a bandwidth extension process to parametrically encode a high band. Further, the method includes generating 2150 a spectral representation of the multi-channel signal and processing 2200 processing the spectral representation including the low-band and high-band of the multi-channel signal to generate multi-channel information.

  FIG. 21 shows a schematic description of a flowchart of a method 2100 for decoding an encoded audio signal, including a core encoded signal, bandwidth extension parameters, and multi-channel information. The method includes decoding 2105 the core-coded signal to generate a monaural signal; converting 2110 the monaural signal to a spectral representation; and extracting the first channel from the spectral representation of the mono signal and the multi-channel information. Generating 2115 a spectrum and a second channel spectrum, synthesizing the first channel spectrum to obtain a first channel signal, and transforming the second channel spectrum to obtain a second channel signal. Synthesizing and filtering 2120.

  Another embodiment is described as follows.

Bitstream Syntax Changes Table 5 of section 5.3.2 Supplementary Payload USAC Standard [1] should be modified as follows.

  The following table should be added.

  The following payload description should be added to Section 6.2, USAC Payload.

6.2. x lpd_stereo_stream ()
The detailed decryption procedure is described in 7. x LPD stereo decoding section.

Terms and definitions lpd_stereo_stream (): Data element for decoding stereo data for LPD mode.
res_mode: a flag indicating the frequency resolution of the parameter band.
q_mode: a flag indicating the time resolution of the parameter band.
ipd_mode: a bit field that defines the maximum value of the parameter band for the IPD parameter.
pred_mode: a flag that indicates if prediction is used.
code_mode: a bit field defining the maximum value of the parameter band for quantizing the side signal.
Ild_idx [k] [b]: ILD parameter index for frame k and band b.
Ipd_idx [k] [b]: IPD parameter index for frame k and band b.
pred_gain_idx [k] [b]: Predicted gain index for frame k and band b.
cod_gain_idx: Global gain index for the quantized side signal.

Auxiliary element ccfl: core code frame length.
M: Table 7. x. Stereo LPD frame length defined in 1.
band_config (): Function to return the number of encoded parameter bands. Function is 7. x.
band_limits (): Function to return the number of encoded parameter bands. Function is 7. x.
max_band (): Function for returning the number of encoded parameter bands. Function is 7. x.
ipd_max_band (): Function to return the number of encoded parameter bands. Function is 7. x.
cod_max_band (): a function to return the number of encoded parameter bands. Function is 7. x.
cod_L: number of DFT lines for the decoded side signal.

Decoding Process LPD Stereo Encoding Tool Description LPD Stereo is a discrete M / S stereo encoding. The intermediate channel is encoded by a monaural LPD core encoder and the side signals are encoded in the DFT domain. The decoded intermediate signal is output from the LPD monaural decoder and then processed by the LPD stereo module. Stereo decoding is done in the DFT domain where the L and R channels are decoded. The two decoded channels are transformed back in the time domain and then combined in this domain with the channels decoded from the FD mode. The FD coding mode uses its own stereo tool, discrete stereo, with or without complex prediction.

Data element res_mode: Flag indicating the frequency resolution of the parameter band.
q_mode: a flag indicating the time resolution of the parameter band.
ipd_mode: a bit field that defines the maximum value of the parameter band for the IPD parameter.
pred_mode: a flag that indicates if prediction is used.
code_mode: a bit field defining the maximum value of the parameter band for quantizing the side signal.
Ild_idx [k] [b]: ILD parameter index for frame k and band b.
Ipd_idx [k] [b]: IPD parameter index for frame k and band b.
pred_gain_idx [k] [b]: Predicted gain index for frame k and band b.
cod_gain_idx: Global gain index for the quantized side signal.

Auxiliary element ccfl: core code frame length.
M: Table 7. x. Stereo LPD frame length defined in 1.
band_config (): Function to return the number of encoded parameter bands. Function is 7. x.
band_limits (): Function to return the number of encoded parameter bands. Function is 7. x.
max_band (): Function for returning the number of encoded parameter bands. Function is 7. x.
ipd_max_band (): Function to return the number of encoded parameter bands. Function is 7. x.
cod_max_band (): a function to return the number of encoded parameter bands. Function is 7. x.
cod_L: number of DFT lines for the decoded side signal.

Decoding process Stereo decoding is performed in the frequency domain. It operates as a post-processing of the LPD decoder. It receives the synthesis of the mono intermediate signal from the LPD decoder. The side signal is then decoded or predicted in the frequency domain. The channel spectrum is then reconstructed in the frequency domain before being resynthesized in the time domain. Stereo LPD works with a fixed frame length equal to the size of the ACELP frame, independent of the encoding mode used in the LPD mode.

Frequency analysis The DFT spectrum of the frame index i is calculated from the decoded frame x of length M.

ここで、Ｎは信号の分析のサイズである。ｗは分析ウィンドウである。ｘは、ＤＦＴのオーバーラップサイズＬにより遅延されたフレームインデックスｉで、ＬＰＤデコーダからの復号化された時間信号である。Ｍは、ＦＤモードの中で使われたサンプリングレートで、ＡＣＥＬＰフレームのサイズと等しい。Ｎは、ステレオＬＰＤフレームサイズおよびＤＦＴのオーバーラップサイズを加えたものと等しい。サイズは、表７．ｘ．１において報告されたように、使われたＬＰＤバージョンに依存している。

Where N is the size of the signal analysis. w is the analysis window. x is the decoded time signal from the LPD decoder, with the frame index i delayed by the overlap size L of the DFT. M is the sampling rate used in the FD mode and is equal to the size of the ACELP frame. N is equal to the sum of the stereo LPD frame size and the DFT overlap size. See Table 7. x. As reported in No. 1, it depends on the LPD version used.

Configuration of Parameter Bands The DFT spectrum is divided into non-overlapping frequency bands called parameter bands. The spectral partitioning is non-uniform, similar to auditory frequency decomposition. Two different divisions of the spectrum are possible with a bandwidth that lasts about twice or about four times the equivalent rectangular bandwidth (ERB). The spectral partitioning is selected by the data element res_mod and is defined by the following pseudo code.

funtion nbands = band_config (N, res_mod)
band_limits [0] = 1;
nbands = 0;
while (band_limits [nbands ++] <(N / 2)) [
if (stereo_lpd_res == 0)
band_limits [nbands] = band_limits_erb2 [nbands];
else
band_limits [nbands] = band_limits_erb4 [nbands];
]
nbands--;
band_limits [nbands] = N / 2;
return nbands

Here, nbands is the total number of parameter bands, and N is the DFT analysis window size. Tables band_limits_erb2 and band_limits_erb4 are shown in Table 7. x. 2 is defined. The decoder can adaptively change the resolution of the spectral parameter band in all two stereo LPD frames.

The maximum number of parameter bands for IPD is sent in a 2-bit field ipd_mod data element.

ipd_max_band = max_band [res_mod] [ipd_mod]

The maximum number of parameter bands for encoding the side signal is sent in a 2-bit field cod_mod data element.

cod_max_band = max_band [res_mod] [cod_mod]

Table max_band [] [] is shown in Table 7. x. 3 is defined.
To predict the side signal, the number of decoded lines is then calculated by the following equation:

cod_L = 2 · (band_limits [cod_max_band] −1)

Dequantization of stereo parameters The stereo parameter inter-channel level difference (ILD), inter-channel phase difference (IPD) and prediction gain are sent to all frames or all two frames depending on the flag q_mode. If q_mode is equal to 0, the parameter updates all frames. Otherwise, the parameter value is updated only for the odd index i of the stereo LPD frame in the USAC frame. The index i of a stereo LPD frame in a USAC frame can be either between 0 and 3 in LPD version 0 and between 0 and 1 in LPD version 1.

The ILD is decoded as follows.

For 0 ≦ b <nbands,
ILD _i [b] = ld_q [ild_idx [i] [b]]

The decoded shape of the side signal is the output of the AVQ described in the USAC standard [1] for the section.

Post Processing Bass post processing is performed separately on the two channels. The process is for both channels, the same as described in section 7.17 of [1].

  It should be understood that in this specification, signals on a line are sometimes named by the reference number of the line, and sometimes are indicated by the reference number itself, which is due to the line. Therefore, the notation is such that a line having a certain signal indicates the signal itself. The line can be a physical line of the implementation connected by wiring. However, in a computerized implementation, there are no physical lines, but the signals represented by the lines are transmitted from one calculation module to another.

  Although the invention has been described in the context of a block diagram, in which the blocks represent actual or logical hardware components, the invention may also be implemented by computer-implemented methods. In the latter case, the blocks represent corresponding method steps, and these steps represent functions performed by the corresponding logical or physical hardware blocks.

  Even if some aspects were described in the context of an apparatus, they will also be understood as representing a corresponding method description. As a result, blocks or devices may correspond to or be understood as features of a method step. By analogy, an aspect has been described with it, or a method step also corresponds to a block, or represents a description of details or characteristics corresponding to an apparatus. Some or all of the method steps may be performed by a hardware device (or using a hardware device), for example, a microprocessor, a programmable computer, or an electronic circuit. In some embodiments, some or some of the most important method steps may be performed by such a device.

  The transmitted or encoded signal of the present invention can be stored on a digital storage medium or transmitted over a transmission medium such as a wireless transmission medium or a wired transmission medium such as the Internet.

  Embodiments of the invention may be implemented in hardware or in software, depending on the particular needs of the implementation. An implementation thereof cooperates with, or cooperates with, a programmable computer system such that a respective method is performed, digital storage having electronically readable control signals stored therein. It may be implemented using a medium, for example, a floppy disk, DVD, Blu-ray disk, CD, ROM, PROM, EPROM, EEPROM, or FLASH memory. Thus, the digital storage medium may be computer readable.

  Some embodiments according to the present invention provide an electronically readable control signal that can cooperate with a programmable computer system such that some of the methods described herein are performed. Data carrier.

  Typically, embodiments of the present invention are embodied as a computer program product having program code, wherein when the computer program product runs on a computer, the program code is activated to perform several methods. . The program code is stored, for example, on a machine-readable carrier.

  Other embodiments include a computer program for performing some of the methods described herein, wherein the computer program is stored on a machine-readable carrier.

  In other words, therefore, when the computer program runs on a computer, an embodiment of the method of the present invention comprises a computer program having program code for performing some of the methods described herein. It is.

  Accordingly, a further embodiment of the method of the present invention is a data carrier (or digital storage medium or computer readable medium) comprising a computer program for performing some of the methods described herein. . The data carrier, digital storage medium or recorded medium is typically tangible and / or intangible.

  Thus, a further embodiment of the method of the invention is a data stream or a series of signals representing a computer program for performing some of the methods described herein. For example, a data stream or series of signals may be configured to be transferred via a data communication connection, eg, the Internet.

  Further embodiments include processing means, eg, a computer, or programmable logic configured or adapted to perform some of the methods described herein.

  Further embodiments include a computer having installed thereon a computer program for performing some of the methods described herein.

  Another embodiment according to the invention includes an apparatus or system configured to transfer a computer program for performing at least one of the methods described herein to a receiver. The transfer is, for example, electronic or optical. The receiver is, for example, a computer or a portable device or a storage device. The device or system includes, for example, a file server for transferring a computer program to a receiver.

  In some embodiments, a programmable logic circuit (eg, a Field Programmable Gate Array (FPGA)) performs some or all of the functions described herein. Can be used for In some embodiments, a field-programmable gate array may cooperate with a microprocessor to perform some of the methods described herein. In general, the method is preferably performed by some hardware device.

  The embodiments described above are merely illustrative of the principles of the present invention. It is understood that modifications and variations of the devices and details described herein will be apparent to others skilled in the art. Thus, it is intended that the discussion of the means and embodiments of the description be limited only by the scope of the following claims, rather than by the detailed description presented herein.

Document [1] ISO / IEC DIS 23003-3, Usac
[2] ISO / IEC DIS 23008-3, 3D Audio

Claims (27)

An audio encoder (2) for encoding a multi-channel signal,
A linear prediction domain encoder (6);
A frequency domain encoder (8);
A controller (10) for switching between the linear prediction domain encoder (6) and the frequency domain encoder (8);
Including
The linear prediction domain encoder (6) includes a downmixer (12) for downmixing the multi-channel signal (4) to obtain a downmix signal (14), and a downmixer (12) for encoding the downmix signal (14). A linear prediction domain core encoder (16), and a first combined multi-channel encoder (18) for generating first multi-channel information (20) from the multi-channel signal;
The frequency domain encoder (8) includes a second combined multi-channel encoder (22) for encoding second multi-channel information (24) from the multi-channel signal, wherein the second combined multi-channel encoder (22) Unlike the first combined multi-channel encoder (18),
The controller (10) is configured such that the portion of the multi-channel signal is represented by either an encoded frame of the linear prediction domain encoder or an encoded frame of the frequency domain encoder. Audio encoder (2).   The first combined multi-channel encoder (18) includes a first time-frequency converter (82); the second combined multi-channel encoder (22) includes a second time-frequency converter (66); The audio encoder (2) according to claim 1, wherein the first time-frequency converter and the second time-frequency converter are different from each other. The first combined multi-channel encoder (18) is a parametric combined multi-channel encoder, or
The audio encoder (2) according to claim 1 or 2, wherein the second combined multi-channel encoder (22) is a waveform maintaining combined multi-channel encoder. The parametric combining multi-channel encoder includes a stereo prediction coder, a parametric stereo encoder or a rotation-based parametric stereo encoder, or
4. The audio encoder of claim 3, wherein the waveform preserving combined multi-channel encoder includes a band selective switch middle / side or left / right stereo coder. The linear prediction domain encoder (6) includes an ACELP processor (30) and a TCX processor (32), wherein the ACELP processor is configured to operate on the downsampled downmix signal (34), A domain bandwidth extension processor (36) configured to parametrically encode a portion of the downmix signal band removed from the ACELP input signal by a third downsampling;
The TCX processor (32) is configured to operate on the unmixed or downsampled downmix signal (14) to a lesser extent than the downsampling for the ACELP processor, the TCX processor comprising: , A first time-frequency converter (40), a first parameter generator (42) for generating a parametric representation (46) of the first band set, and a quantized encoder for the second band set. An audio encoder according to any of the preceding claims, comprising a first quantization encoder (44) for generating a set of spectral lines (48).   The frequency domain encoder (8) converts a first channel (4a) of the multi-channel signal (4) and a second channel (4b) of the multi-channel signal (4) into a spectral representation (72a, 72b). A second time-frequency converter (66), a second parameter generator (68) for generating a parametric representation of the second band set, and a quantized and encoded first band set (80). An audio encoder (2) according to any of the preceding claims, including a second quantization encoder (70) for generating a rendered representation. The linear prediction domain encoder includes an ACELP processor with time domain bandwidth extension, a TCX processor with MDCT operation, and an intelligent gap filling function, or
The frequency domain encoder includes an MDCT operation for the first channel and the second channel, an AAC operation, and an intelligent gap filling function, or
7. The first combined multi-channel encoder is configured to operate in such a way that multi-channel information for the entire bandwidth of the multi-channel audio signal is derived. 8. (2). A linear prediction domain decoder (50) for decoding the downmix signal (14) to obtain an encoded and decoded downmix signal (54);
The encoded and decoded downmix signal (54) representing an error between the decoded multichannel representation using the first multichannel information (20) and the multichannel signal before downmixing. ), A multi-channel residual coder (56) for calculating and encoding the multi-channel residual signal (58);
Audio encoder (2) according to any of the preceding claims, further comprising:   The downmix signal has a low band and a high band, the linear prediction domain encoder is configured to apply a bandwidth extension process to parametrically encode the high band, and the linear prediction domain decoder , The encoded multi-channel residual signal is configured to obtain only a low-band signal representing the low band of the down-mix signal as the encoded and decoded down-mix signal (54). The audio encoder (2) according to claim 8, wherein (58) has only a frequency within the low band of the multi-channel signal before downmixing. The multi-channel residual coder (56)
A combined multi-channel decoder (60) for generating a decoded multi-channel signal (64) using the first multi-channel information (20) and the encoded and decoded downmix signal (54). )When,
A difference processor (62) for forming a difference between the decoded multi-channel signal and the multi-channel signal before downmixing to obtain the multi-channel residual signal;
Audio encoder (2) according to claim 8 or claim 9, comprising: The downmixer (12) is configured to convert the multi-channel signal into a spectral representation, wherein the downmixing is performed using the spectral representation or using a time domain representation;
The first multi-channel encoder is configured to use the spectral representation to generate separate first multi-channel information for individual bands of the spectral representation. Audio encoder (2) according to any one of the above. The controller (10) uses a frequency domain encoder (8) for encoding the previous frame in the current frame (204) of the multi-channel audio signal, so that it can decode future frames. Configured to switch to the linear prediction domain encoder,
The first combined multi-channel encoder (18) is configured to calculate a composite multi-channel parameter (210a, 210b, 212a, 212b) from the multi-channel audio signal for the current frame;
The audio encoder (2) according to any of the preceding claims, wherein the second combined multi-channel encoder (22) is configured to weight the second multi-channel signal using a stop window. An audio decoder (102) for decoding an encoded audio signal (103),
A linear prediction domain decoder (104);
A frequency domain decoder (106);
A first combined multi-channel decoder (108) for generating a first multi-channel representation (114) using an output of the linear prediction domain decoder (104) and first multi-channel information (20);
A second combined multi-channel decoder (110) for generating a second multi-channel representation (116) using an output of the frequency domain decoder (106) and second multi-channel information (22, 24);
A first combiner (112) for combining the first multi-channel representation (114) and the second multi-channel representation (116) to obtain a decoded audio signal (118);
The second combined multi-channel decoder is an audio decoder different from the first combined multi-channel decoder. The first combined multi-channel decoder is a parametric combined multi-channel decoder, the second combined multi-channel decoder is a waveform maintaining combined multi-channel decoder,
The first combined multi-channel decoder is configured to operate based on a complex prediction, a parametric stereo operation or a rotation operation;
The audio decoder (102) of claim 13, wherein the second combined multi-channel decoder is configured to apply a band selective switch to a middle / side or left / right stereo decoding algorithm. The linear prediction domain decoder,
An ACELP decoder (120), a low-band synthesizer (122), an up-sampler (124), a time-domain bandwidth extension processor (126), or for combining the up-sampled signal with the bandwidth-extended signal. A second coupler (128),
A TCX decoder (130) and an intelligent gap filling processor (132);
A full-band synthesis processor (134) for combining the output of the second combiner (128) with the outputs of the TCX decoder (130) and the IGF processor (132); or
A cross-path (136) is provided for initializing the low-band synthesizer using information derived by low-band spectral time conversion from the TCX decoder and the IGF processor. An audio decoder (102) as described. A time-frequency converter (138) for converting the output of the linear prediction domain decoder (104) into a spectral representation (145);
An up-mixer controlled by the first multi-channel information acting on the spectral representation (145);
16. The audio decoder (102) according to any of claims 13 to 15, including a frequency-to-time converter (148) for converting the upmix result into a time representation period. The second combined multi-channel decoder (110) is configured to use as input a spectral representation obtained by the frequency domain decoder, the spectral representation comprising, for at least a plurality of bands, a first channel signal and a second Including channel signals,
Applying a combined multi-channel operation to a plurality of bands of the first channel signal and the second channel signal, converting the result of the combined multi-channel decoder combined multi-channel operation into a time representation to obtain the second multi-channel representation. An audio decoder (102) according to any of claims 13 to 16, configured as follows.   The second multi-channel information (22) is a mask indicating left / right or middle / side combined multi-channel coding for each band, and the combined multi-channel operation is performed in accordance with the band indicated by the mask. Audio decoder (102) according to claim 17, which is an intermediate / side to left / right conversion operation for converting from the intermediate / side representation to a left / right representation. The encoded multi-channel audio signal includes a residual signal for the output of the linear prediction domain decoder;
19. The audio decoder (10) according to any of claims 13 to 18, wherein the first combined multi-channel decoder is configured to use the multi-channel residual signal to generate the first multi-channel representation. 102).   The multi-channel residual signal has a lower bandwidth than the first multi-channel representation, and the first combined multi-channel decoder reconstructs an intermediate first multi-channel representation using the first combined multi-channel information. 20. The audio decoder (102) of claim 19, wherein the audio decoder (102) is configured and configured to add the multi-channel residual signal to the intermediate first multi-channel representation. The time-frequency converter includes a complex operation or an oversampling operation;
The audio decoder (102) of claim 16, wherein the frequency domain decoder comprises an IMDCT operation or a critical sampling operation. The audio decoder (102) decodes a subsequent frame in the current frame (204) of the multi-channel audio signal by using the frequency domain decoder (106) to decode a previous frame. The linear prediction domain decoder (104) for
The combiner (112) is configured to calculate a composite intermediate signal (226) from the second multi-channel representation (116) of the current frame;
The first combined multi-channel decoder (108) is configured to generate the first multi-channel representation (114) using the composite intermediate signal (226) and first multi-channel information (20);
14. The combiner (112) is configured to combine the first multi-channel representation and the second multi-channel representation to obtain a decoded current frame of the multi-channel audio signal. An audio decoder (102) according to any of the preceding claims. The audio decoder (102) decodes subsequent frames in the current frame (232) of the multi-channel audio signal by using the linear prediction domain decoder (104) to decode previous frames. Switching to the frequency domain decoder (106) for performing
The stereo decoder (146) is configured to calculate a composite multi-channel audio signal from the decoded mono signal of the linear prediction domain decoder for a current frame using multi-channel information of a previous frame. ,
The second combined multi-channel decoder (110) is configured to calculate a second multi-channel representation for the current frame and to weight the second multi-channel representation using a start window;
The combiner (112) is configured to combine the composite multi-channel audio signal and the weighted second multi-channel representation to obtain a decoded current frame of the multi-channel audio signal. An audio decoder (102) according to any of claims 13 to 22.   The audio encoder according to any one of claims 1 to 12, or the audio decoder according to any one of claims 13 to 23, wherein the multi-channel means two or more channels. A method (800) for encoding a multi-channel signal, comprising:
Performing a linear prediction domain encoding;
Performing frequency domain coding;
Switching between the linear prediction domain encoding and the frequency domain encoding;
Including
The step of performing the linear prediction domain coding includes a downmix signal, a linear prediction domain for core-coding the downmix signal, and a first combined multipath for generating first multichannel information from the multichannel signal. Downmixing the multi-channel signal to obtain a channel encoding,
The performing the frequency domain coding includes performing a second combined multi-channel encoding that generates second multi-channel information from the multi-channel signal, and the performing the second combined multi-channel encoding includes a first Unlike the step of multi-channel encoding,
The switching is performed such that the portion of the multi-channel signal is represented by either a coded frame of the linear prediction domain coding or a coded frame of the frequency domain coding. The method (800). A method (900) for decoding an encoded audio signal, comprising:
Decoding a linear prediction domain;
Frequency domain decoding;
Performing a first combined multi-channel decoding to generate a first multi-channel representation using an output of the linear prediction domain decoding and first multi-channel information;
Performing a second combined multi-channel decoding to generate a second multi-channel representation using the output of the frequency domain decoding and the second multi-channel information;
Combining the first multi-channel representation and the second multi-channel representation to obtain a decoded audio signal;
Including
The method of decoding an encoded audio signal, wherein the step of decoding the second multi-channel is different from the step of decoding the first multi-channel (900).   A computer program for performing the method of claim 25 or claim 26 when the computer program runs on a computer or processor.

Images